ILBP算子在浓雾天气形势图识别中的应用
|
摘要
纹理是天气形势图的突出特征,有效地从天气形势图提取并表示其纹理是实现雾型实时在线预报的基础。基于此,提出一种改进的局部二值模式算法,通过调整局部二值模式(LBP)算子中二进制多项式的权重,实现其提取特定方向上纹理特征的目标。将江苏地区2010年—2017年500张浓雾天气形势图作为数据集,采用Chi统计法匹配测试数据与基准数据的相似度进行天气分类。实验结果表明,该算法的准确率、虚警率及临界成功指数分别为0.884、0.15和0.76,均优于LBP算法,具有较高的识别准确性与可靠性。
Texture is the salient feature of a weather situation map.Effectively extracting the textures from the weather situation map and representing them is the basis for real-time online prediction of fog pattern.To this end,this paper proposes an Improved Local Binary Pattern(ILBP) algorithm,which can extract texture features in a specific direction by adjusting the weight coefficients of binary polynomials in Local Binary Mode(LBP) operators.500 fog weather situation maps in Jiangsu Province from 2010 to 2017 are used as the dataset,and the similarities between the test data and the benchmark data are matched by Chi statistical method to classify the weather.Experimental results show that,the Probability of Detection(PoD),False Alarm Rate(FAR) and Critical Success Index(CSI) of the proposed algorithm are 0.884,0.15 and 0.76,respectively,which are better than those of the LBP algorithm.It proves that the proposed model has high accuracy and reliability.
Texture is the salient feature of a weather situation map.Effectively extracting the textures from the weather situation map and representing them is the basis for real-time online prediction of fog pattern.To this end,this paper proposes an Improved Local Binary Pattern(ILBP) algorithm,which can extract texture features in a specific direction by adjusting the weight coefficients of binary polynomials in Local Binary Mode(LBP) operators.500 fog weather situation maps in Jiangsu Province from 2010 to 2017 are used as the dataset,and the similarities between the test data and the benchmark data are matched by Chi statistical method to classify the weather.Experimental results show that,the Probability of Detection(PoD),False Alarm Rate(FAR) and Critical Success Index(CSI) of the proposed algorithm are 0.884,0.15 and 0.76,respectively,which are better than those of the LBP algorithm.It proves that the proposed model has high accuracy and reliability.
引文
[1] 李子华.中国近40年来雾的研究[J].气象学报,2001,59(5):616-624.
[2] 牛生杰,陆春松,吕晶晶,等.近年来中国雾研究进展[J].气象科技进展,2016,6(2):6-19.
[3] 孙玉,刘贵全.基于HOG与LBP特征的人脸识别方法[J].计算机工程,2015,41(9):205-208,214.
[4] 刘丽,匡纲要.图像纹理特征提取方法综述[J].中国图象图形学报,2009,14(4):622-635.
[5] LIU Guanghai,YANG Jingyu.Image retrieval based on the texton co-occurrence matrix[J].Pattern Recognition,2008,41(12):3521-3527.
[6] KASETKASEM T,ARORA M K,VARSHNEY P K.Super-resolution land cover mapping using a Markov random field based approach[J].Remote Sensing of Environment,2005,96(3/4):302-314.
[7] ARIVAZHAGAN S,GANESAN L.Texture classification using wavelet transform[J].Pattern Recognition Letters,2003,24(9/10):1513-1521.
[8] 袁紫华,李峰,周书仁.基于Haar型LBP纹理特征的人体姿态估计[J].计算机工程,2015,41(4):199-204.
[9] 宋本钦,李培军.加入改进LBP纹理的高分辨率遥感图像分类[J].国土资源遥感,2010(4):40-45.
[10] 徐先传,张琦.基于LBP算子的医学图像检索方法[J].微计算机信息,2007,23(18):281-282.
[11] 瞿中,张亢,乔高元,等.MB-LBP特征提取和粒子滤波相结合的运动目标检测与跟踪算法研究[J].计算机科学,2013,40(12):304-307.
[12] 吴俊,李文杰,耿磊,等.其于单目视觉的前方车辆检测与测距[J].计算机工程,2017,43(2):26-32.
[13] 徐胜军,刘欣,赵亮.基于快速收敛LBP算法的图像分割[J].计算机应用,2011,31(8):2229-2231.
[14] 李艳玮,郑伟勇,林楠.融合AAM、CNN与LBP特征的人脸表情识别方法[J].计算机工程与设计,2017(12):254-258.
[15] WEI Xia,YIN Souyi,OUYANG Peng.A high precision feature based on LBP and Gabor theory for face recognition[J].Sensors,2013,13(4):4499-4513.
[16] 黄非非.基于LBP的人脸识别研究[D].重庆:重庆大学,2009.
[17] 张伟康,马慧云,邹峥嵘,等.基于SBDART辐射传输模型的夜间辐射雾自动检测及时间序列分析[J].国土资源遥感,2014,26(2):80-86.
[2] 牛生杰,陆春松,吕晶晶,等.近年来中国雾研究进展[J].气象科技进展,2016,6(2):6-19.
[3] 孙玉,刘贵全.基于HOG与LBP特征的人脸识别方法[J].计算机工程,2015,41(9):205-208,214.
[4] 刘丽,匡纲要.图像纹理特征提取方法综述[J].中国图象图形学报,2009,14(4):622-635.
[5] LIU Guanghai,YANG Jingyu.Image retrieval based on the texton co-occurrence matrix[J].Pattern Recognition,2008,41(12):3521-3527.
[6] KASETKASEM T,ARORA M K,VARSHNEY P K.Super-resolution land cover mapping using a Markov random field based approach[J].Remote Sensing of Environment,2005,96(3/4):302-314.
[7] ARIVAZHAGAN S,GANESAN L.Texture classification using wavelet transform[J].Pattern Recognition Letters,2003,24(9/10):1513-1521.
[8] 袁紫华,李峰,周书仁.基于Haar型LBP纹理特征的人体姿态估计[J].计算机工程,2015,41(4):199-204.
[9] 宋本钦,李培军.加入改进LBP纹理的高分辨率遥感图像分类[J].国土资源遥感,2010(4):40-45.
[10] 徐先传,张琦.基于LBP算子的医学图像检索方法[J].微计算机信息,2007,23(18):281-282.
[11] 瞿中,张亢,乔高元,等.MB-LBP特征提取和粒子滤波相结合的运动目标检测与跟踪算法研究[J].计算机科学,2013,40(12):304-307.
[12] 吴俊,李文杰,耿磊,等.其于单目视觉的前方车辆检测与测距[J].计算机工程,2017,43(2):26-32.
[13] 徐胜军,刘欣,赵亮.基于快速收敛LBP算法的图像分割[J].计算机应用,2011,31(8):2229-2231.
[14] 李艳玮,郑伟勇,林楠.融合AAM、CNN与LBP特征的人脸表情识别方法[J].计算机工程与设计,2017(12):254-258.
[15] WEI Xia,YIN Souyi,OUYANG Peng.A high precision feature based on LBP and Gabor theory for face recognition[J].Sensors,2013,13(4):4499-4513.
[16] 黄非非.基于LBP的人脸识别研究[D].重庆:重庆大学,2009.
[17] 张伟康,马慧云,邹峥嵘,等.基于SBDART辐射传输模型的夜间辐射雾自动检测及时间序列分析[J].国土资源遥感,2014,26(2):80-86.